Data mining crystallization databases: knowledge-based approaches to optimize protein crystal screens

Protein crystallization is a major bottleneck in protein X-ray crystallography, the workhorse of most structural proteomics projects. Because the principles that govern protein crystallization are too poorly understood to allow them to be used in a strongly predictive sense, the most common crystallization strategy entails screening a wide variety of solution conditions to identify the small subset that will support crystal nucleation and growth. We tested the hypothesis that more efficient crystallization strategies could be formulated by extracting useful patterns and correlations from the large data sets of crystallization trials created in structural proteomics projects. A database of crystallization conditions was constructed for 755 different proteins purified and crystallized under uniform conditions. Forty-five percent of the proteins formed crystals. Data mining identified the conditions that crystallize the most proteins, revealed that many conditions are highly correlated in their behavior, and showed that the crystallization success rate is markedly dependent on the organism from which proteins derive. Of the proteins that crystallized in a 48-condition experiment, 60% could be crystallized in as few as 6 conditions and 94% in 24 conditions. Consideration of the full range of information coming from crystal screening trials allows one to design screens that are maximally productive while consuming minimal resources, and also suggests further useful conditions for extending existing screens.     

Automation aided optimization of cloning,expression and purification of enzymes of the bacterial sialic acid catabolic and sialylation pathways enzymes for structural studies

The process of obtaining a well‐expressing, soluble and correctly folded constructs can be made easier and quicker by automating the optimization of cloning, expression and purification. While there are many semiautomated pipelines available for cloning, expression and purification, there is hardly any pipeline that involves complete automation. Here, we achieve complete automation of all the steps involved in cloning and in vivo expression screening. This is demonstrated using 18 genes involved in sialic acid catabolism and the surface sialylation pathway. Our main objective was to clone these genes into a His‐tagged Gateway vector, followed by their small‐scale expression optimization in vivo. The constructs that showed best soluble expression were then selected for purification studies and scaled up for crystallization studies. Our technique allowed us to quickly find conditions for producing significant quantities of soluble proteins in Escherichia coli, their large‐scale purification and successful crystallization of a number of these proteins. The method can be implemented in other cases where one needs to screen a large number of constructs, clones and expression vectors for successful recombinant production of functional proteins.     

Functional and structural genomics using PEDANT

MOTIVATION: Enormous demand for fast and accurate analysis of biological sequences is fuelled by the pace of genome analysis efforts. There is also an acute need in reliable up-to-date genomic databases integrating both functional and structural information. Here we describe the current status of the PEDANT software system for high-throughput analysis of large biological sequence sets and the genome analysis server associated with it. RESULTS: The principal features of PEDANT are: (i) completely automatic processing of data using a wide range of bioinformatics methods, (ii) manual refinement of annotation, (iii) automatic and manual assignment of gene products to a number of functional and structural categories, (iv) extensive hyperlinked protein reports, and (v) advanced DNA and protein viewers. The system is easily extensible and allows to include custom methods, databases, and categories with minimal or no programming effort. PEDANT is actively used as a collaborative environment to support several on-going genome sequencing projects. The main purpose of the PEDANT genome database is to quickly disseminate well-organized information on completely sequenced and unfinished genomes. It currently includes 80 genomic sequences and in many cases serves as the only source of exhaustive information on a given genome. The database also acts as a vehicle for a number of research projects in bioinformatics. Using SQL queries, it is possible to correlate a large variety of pre-computed properties of gene products encoded in complete genomes with each other and compare them with data sets of special scientific interest. In particular, the availability of structural predictions for over 300 000 genomic proteins makes PEDANT the most extensive structural genomics resource available on the web.     

Randomness in crystallization of proteins from Staphylococcus aureus

Of many factors affecting protein crystallization, randomness in proteins has been given less attention although highly structured proteins would be at low entropy state. The factors, which impact on protein crystallization, are almost exclusively related to non-random amino acid properties such as physiochemical properties of amino acids. In this study, we used logistic regression and neural network to model the success rate of crystallization of 420 proteins from Staphylococcus aureus with each of non-random and random amino acid properties in order to determine whether randomness in a protein plays a role in the crystallization process. The results show that randomness is indeed involved in the crystallization process, and this rationale would enrich our knowledge on crystallization process and enhance our ability to crystallize more important proteins.     

Cellulose biosynthesis: current views and evolving concepts

* AIMS: To outline the current state of knowledge and discuss the evolution of various viewpoints put forth to explain the mechanism of cellulose biosynthesis. * SCOPE: Understanding the mechanism of cellulose biosynthesis is one of the major challenges in plant biology. The simplicity in the chemical structure of cellulose belies the complexities that are associated with the synthesis and assembly of this polysaccharide. Assembly of cellulose microfibrils in most organisms is visualized as a multi-step process involving a number of proteins with the key protein being the cellulose synthase catalytic sub-unit. Although genes encoding this protein have been identified in almost all cellulose synthesizing organisms, it has been a challenge in general, and more specifically in vascular plants, to demonstrate cellulose synthase activity in vitro. The assembly of glucan chains into cellulose microfibrils of specific dimensions, viewed as a spontaneous process, necessitates the assembly of synthesizing sites unique to most groups of organisms. The steps of polymerization (requiring the specific arrangement and activity of the cellulose synthase catalytic sub-units) and crystallization (directed self-assembly of glucan chains) are certainly interlinked in the formation of cellulose microfibrils. Mutants affected in cellulose biosynthesis have been identified in vascular plants. Studies on these mutants and herbicide-treated plants suggest an interesting link between the steps of polymerization and crystallization during cellulose biosynthesis. * CONCLUSIONS: With the identification of a large number of genes encoding cellulose synthases and cellulose synthase-like proteins in vascular plants and the supposed role of a number of other proteins in cellulose biosynthesis, a complete understanding of this process will necessitate a wider variety of research tools and approaches than was thought to be required a few years back.     

Molecular mechanism for the crystallization of bacteriorhodopsin in lipidic cubic phases

Crystals of transmembrane proteins may be grown from detergent solutions or in a matrix of membranous lipid bilayers existing in a liquid crystalline state and forming a cubic phase (in cubo). While crystallization in micellar solutions appears analogous to that for soluble proteins, crystallization in lipidic matrices is poorly understood. As this method was shown to be applicable to several membrane proteins, understanding its mechanism will facilitate a rational design of crystallization, minimizing the laborious screening of a large number of parameters. Using polarization microscopy and low-angle X-ray diffraction, experimental evidence is provided to support a mechanistic model for the in cubo crystallization of bacteriorhodopsin in a lipid matrix. Membrane proteins are thought to reside in curved lipid bilayers, to diffuse into patches of lower curvature and to incorporate into lattices which associate to form highly ordered three-dimensional crystals. Critical testing of this model is necessary to generalize it to other membrane proteins.     

Identification of novel homologues of three low molecular weight subunits of the mitochondrial bc1 complex

Large-scale random cDNA sequencing projects have been started for several organisms and are a valuable tool for the analysis of quantitative and qualitative aspects of gene expression. However, the reliability of the obtained data is limited as most of the clones are only partially analysed on one strand. As a consequence the sequence entries derived from random cDNA sequencing projects usually comprise incomplete open reading frames. They nevertheless define complete and reliable coding sequences, if two prerequisites are fullfilled: (i) the clones encode very small proteins, and (ii) the clones have a high frequency in the cDNA-banks. The present study describes the use of cDNA databases for the identification of homologues of three low-molecular-weight subunits of the mitochondrial bc₁ complex, termed the QCR6, QCR9 and QCR10 proteins. These polypeptides are only characterized for a small number of organisms, have a scarcely defined function and exhibit a low degree of structural conservation if compared between different species. Several clones were identified for each polypeptide by searches with TBLASTN using the known sequences as probes. Most of the database entries contain complete open reading frames and sequencing queries could be excluded due to the abundancy of the clones. Multiple sequence alignments are presented for all three polypeptides and consensus sequences are given which may provide a basis for the investigation of the proteins by site-directed mutagenesis.     

Automated interpretation of subcellular patterns in fluorescence microscope images for location proteomics.

Proteomics, the large scale identification and characterization of many or all proteins expressed in a given cell type, has become a major area of biological research. In addition to information on protein sequence, structure and expression levels, knowledge of a protein's subcellular location is essential to a complete understanding of its functions. Currently, subcellular location patterns are routinely determined by visual inspection of fluorescence microscope images. We review here research aimed at creating systems for automated, systematic determination of location. These employ numerical feature extraction from images, feature reduction to identify the most useful features, and various supervised learning (classification) and unsupervised learning (clustering) methods. These methods have been shown to perform significantly better than human interpretation of the same images. When coupled with technologies for tagging large numbers of proteins and high-throughput microscope systems, the computational methods reviewed here enable the new subfield of location proteomics. This subfield will make critical contributions in two related areas. First, it will provide structured, high-resolution information on location to enable Systems Biology efforts to simulate cell behavior from the gene level on up. Second, it will provide tools for Cytomics projects aimed at characterizing the behaviors of all cell types before, during, and after the onset of various diseases.     

Rapid Isolation of Single-chain Antibodies for Structural Genomics

High throughput approaches to structural genomics requires expression, purification, and crystallization of proteins derived from predicted open reading frames cloned into a host organism, typically E. coli. Early results from this approach suggest that the success rate of obtaining well diffracting crystals from eukaryotic proteins is disappointingly low. A proven method of improving the odds of crystallization is formation of a complex with a conformation-stabilizing partner of known structure that is easily crystallized. Such complexes are also able to engage in different crystal contacts than the original protein by itself. Fab fragments derived from monoclonal antibodies have been successfully used for this purpose for a variety of proteins, however conventional methods for the isolation of monoclonal antibodies from hybridomas are time consuming and expensive. We are exploring the use of phage display to generate recombinant antibodies to target proteins that can be used to obtain co-complexes to facilitate crystallization and structural determination. We are using a large, human single-chain Fv (scFv) library to select for antibodies that bind to a panel of Leishmania major target proteins. Thirteen out of 16 target proteins yielded good binders after three rounds of enrichment. A total of 55 distinct scFvs were identified, with five targets each yielding at least five different scFvs. Individual clones were analyzed for binding specificity and soluble scFv can be readily produced and purified via the appended His₆ epitope tag. Using immunoaffinity chromatography, eight scFv target protein pairs were identified that exhibit stable complex formation and are suitable for co-crystallization trials.     

The crystallization of biological macromolecules under microgravity: a way to more accurate three-dimensional structures?

The crystallization of proteins and other biological particles (including nucleic acids, nucleo-protein complexes and large assemblies such as nucleosomes, ribosomal subunits or viruses) in a microgravity environment can produce crystals having lesser defects than crystals prepared under normal gravity on earth. Such microgravity-grown crystals can diffract X-rays to a higher resolution and have a lower mosaic spread. The inferred electron density maps can be richer in details owing to which more accurate three-dimensional structure models can be built. Major results reported in this field of research are reviewed. Novel ones obtained with the Advanced Protein Crystallization Facility are presented. For structural biology, practical applications and implications associated with crystallization and crystallography onboard the International Space Station are discussed.     

Two-dimensional crystallization of membrane proteins: the lipid layer strategy

Due to the difficulty to crystallize membrane proteins, there is a considerable interest to intensify research topics aimed at developing new methods of crystallization. In this context, the lipid layer crystallization at the air/water interface, used so far for soluble proteins, has been recently adapted successfully to produce two-dimensional (2D) crystals of membrane proteins, amenable to structural analysis by electron crystallography. Besides to represent a new alternative strategy, this approach gains the advantage to decrease significantly the amount of material needed in incubation trials, thus opening the field of crystallization to those membrane proteins difficult to surexpress and/or purify. The systematic studies that have been performed on different classes of membrane proteins are reviewed and the physico-chemical processes that lead to the production of 2D crystals are addressed. The different drawbacks, advantages and perspectives of this new strategy for providing structural information on membrane proteins are discussed.     

Tools for visualization and integration of intermediate sequencing results in large disease gene discovery projects

We describe two Java applets which are useful for insightful presentation of intermediate experimental data in gene discovery projects involving large scale sequencing. One of these applets provides a physical map of a genomic region and provides easy access to the second applet, which furnishes a detailed map of sequence contigs associated with clones on the physical map. In particular, the second applet displays all the known information about each contig, including the presence of exons, database homology `hits', repetitive elements and other features; the graphics are linked to other World Wide Web pages, providing detailed information on each feature. These applets should be useful to other research groups working on large sequencing projects.     

Macro-to-micro structural proteomics: native source proteins for high-throughput crystallization

Structural biology and structural genomics projects routinely rely on recombinantly expressed proteins, but many proteins and complexes are difficult to obtain by this approach. We investigated native source proteins for high-throughput protein crystallography applications. The Escherichia coli proteome was fractionated, purified, crystallized, and structurally characterized. Macro-scale fermentation and fractionation were used to subdivide the soluble proteome into 408 unique fractions of which 295 fractions yielded crystals in microfluidic crystallization chips. Of the 295 crystals, 152 were selected for optimization, diffraction screening, and data collection. Twenty-three structures were determined, four of which were novel. This study demonstrates the utility of native source proteins for high-throughput crystallography.     

The Proteome Analysis database: a tool for the in silico analysis of whole proteomes

The Proteome Analysis database (http://www.ebi.ac.uk/proteome/) has been developed by the Sequence Database Group at EBI utilizing existing resources and providing comparative analysis of the predicted protein coding sequences of the complete genomes of bacteria, archeae and eukaryotes. Three main projects are used, InterPro, CluSTr and GO Slim, to give an overview on families, domains, sites, and functions of the proteins from each of the complete genomes. Complete proteome analysis is available for a total of 89 proteome sets. A specifically designed application enables InterPro proteome comparisons for any one proteome against any other one or more of the proteomes in the database.     

Systematic use of the incomplete factorial approach in the design of protein crystallization experiments

We have used the Incomplete Factorial Approach (Carter, C. W., and Carter, C. W., Jr. (1979) J. Biol. Chem. 254, 12219-12223) in conjunction with the program Cristal (Roussel, A., Serre, L., Frey M., and Fontecilla-Camps, J. (1990) J. Crystal Growth 106, 405-409) to crystallize six different proteins. We were able to obtain crystals and to identify the critical factors for crystallization for each of these six proteins. In some of the cases, we succeeded on the first try while using only minute amounts of protein. This study proves that the Incomplete Factorial Approach is a powerful tool in identifying the factors that need to be varied to achieve crystallization. Single crystals of adequate size were obtained for all the proteins reported here, although some did not diffract well enough to be studied by x-ray diffraction methods; the asymmetric units of these latter crystals contain a large metric units of these latter crystals contain a large number of molecules, which is most likely due to the presence of significant amounts of carbohydrate in the proteins.     

Computational Prediction of Protein–Protein Interactions

Recently a number of computational approaches have been developed for the prediction of protein–protein interactions. Complete genome sequencing projects have provided the vast amount of information needed for these analyses. These methods utilize the structural, genomic, and biological context of proteins and genes in complete genomes to predict protein interaction networks and functional linkages between proteins. Given that experimental techniques remain expensive, time-consuming, and labor-intensive, these methods represent an important advance in proteomics. Some of these approaches utilize sequence data alone to predict interactions, while others combine multiple computational and experimental datasets to accurately build protein interaction maps for complete genomes. These methods represent a complementary approach to current high-throughput projects whose aim is to delineate protein interaction maps in complete genomes. We will describe a number of computational protocols for protein interaction prediction based on the structural, genomic, and biological context of proteins in complete genomes, and detail methods for protein interaction network visualization and analysis.     

An improved tripod amphiphile for membrane protein solubilization

Intrinsic membrane proteins represent a large fraction of the proteins produced by living organisms and perform many crucial functions. Structural and functional characterization of membrane proteins generally requires that they be extracted from the native lipid bilayer and solubilized with a small synthetic amphiphile, for example, a detergent. We describe the development of a small molecule with a distinctive amphiphilic architecture, a "tripod amphiphile," that solubilizes both bacteriorhodopsin (BR) and bovine rhodopsin (Rho). The polar portion of this amphiphile contains an amide and an amine-oxide; small variations in this polar segment are found to have profound effects on protein solubilization properties. The optimal tripod amphiphile extracts both BR and Rho from the native membrane environments and maintains each protein in a monomeric native-like form for several weeks after delipidation. Tripod amphiphiles are designed to display greater conformational rigidity than conventional detergents, with the long-range goal of promoting membrane protein crystallization. The results reported here represent an important step toward that ultimate goal.     

Thermofluor-based high-throughput stability optimization of proteins for structural studies

Production of proteins well suited for structural studies is inherently difficult and time-consuming. Protein sample homogeneity, stability, and solubility are strongly correlated with the proteins' probability of yielding crystals, and optimization of these properties will improve success rates of crystallization. In the current study, we applied the thermofluor method as a high-throughput approach for identifying optimal protein formulation for crystallization. The method also allowed optimal stabilizing buffer compositions to be rapidly identified for each protein. Furthermore, the method allowed the identification of potential ligands, physiological or non-physiological, that can be used in subsequent crystallization trials. For this study, the thermally induced melting points were determined in different buffers as well as with additives for a total of 25 Escherichia coli proteins. Crystallization trials were set up together with stabilizing and destabilizing additives identified using thermofluor screening. A twofold increase in the number of crystallization leads was observed when the proteins were cocrystallized with stabilizing additives as compared with experiments without these additives. This suggests that thermofluor constitutes an efficient generic high-throughput method for identification of protein properties predictive of crystallizability.     

Proteome Analysis Database: online application of InterPro and CluSTr for the functional classification of proteins in whole genomes

The SWISS-PROT group at EBI has developed the Proteome Analysis Database utilising existing resources and providing comparative analysis of the predicted protein coding sequences of the complete genomes of bacteria, archaea and eukaryotes (http://www.ebi.ac. uk/proteome/). The two main projects used, InterPro and CluSTr, give a new perspective on families, domains and sites and cover 31-67% (InterPro statistics) of the proteins from each of the complete genomes. CluSTr covers the three complete eukaryotic genomes and the incomplete human genome data. The Proteome Analysis Database is accompanied by a program that has been designed to carry out InterPro proteome comparisons for any one proteome against any other one or more of the proteomes in the database.     

Protein Production and Crystallization at SECSG – An Overview

Using a high degree of automation, the Southeast Collaboratory for Structural Genomics (SECSG) has developed high throughput pipelines for protein production, and crystallization using a two-tiered approach. Primary, or tier-1, protein production focuses on producing proteins for members of large Pfam families that lack a representative structure in the Protein Data Bank. Target genomes are Pyrococcus furiosus and Caenorhabditis elegans. Selected human proteins are also under study. Tier-2 protein production, or target rescue, focuses on those tier-1 proteins, which either fail to crystallize or give poorly diffracting crystals. This two tier approach is more efficient since it allows the primary protein production groups to focus on the production of new targets while the tier-2 efforts focus on providing additional sample for further studies and modified protein for structure determination. Both efforts feed the SECSG high throughput crystallization pipeline, which is capable of screening over 40 proteins per week. Details of the various pipelines in use by the SECSG for protein production and crystallization, as well as some examples of target rescue are described.